Closed-Loop Molecular Design with Calibrated Deference
For AI-driven molecular design, CLIO demonstrates the ability to recognize and adapt to failures of its own predictive models, enabling robust closed-loop optimization in complex experimental settings.
CLIO, a reasoning agent with calibrated deference, guided a closed-loop human-AI campaign to design an aqueous organic redox flow battery negolyte, achieving a 130 mV improvement in redox potential over baseline and then diagnosing and correcting an unexpected reversibility issue, ultimately delivering a compound with 90 mV improvement and improved reversibility.
We present Cognitive Loop via In-Situ Optimization (CLIO), an agent that couples a continuously-updated belief-state graph with a recursive plan-then-act loop. The result is a reasoning agent that can contribute something qualitatively different, which we term \emph{calibrated deference}: the capacity to recognize when its own tools or assumptions are failing, to adapt its strategy in response, and to generate mechanistic hypotheses that guide experimental revision. We tested CLIO in a closed-loop human-AI campaign to design an aqueous organic redox flow battery (AORFB) negolyte, with CLIO leading proposal and interpretation in close partnership with chemists who synthesized, characterized, and weighed in on design choices. Across 17 candidates over three rounds, CLIO converged on a top phosphonate candidate; characterization confirmed a 130~mV improvement in redox potential over the literature baseline. Characterization then revealed unexpectedly poor electrochemical reversibility -- a regression no property predictor had flagged. CLIO generated competing mechanistic hypotheses, prioritized discriminating diagnostics, traced the failure to phosphonate-potassium ion pairing, and prescribed a sulfonate replacement. The resulting compound showed substantially improved electrochemical reversibility and maintained a 90~mV improvement in redox potential, closing the design-make-test-redesign loop.